4 min read

Innovation in Times of Crisis

Originally published in Chicago Medical Magazine

This past week's policy guidance by the U.S. Food and Drug Administration (FDA) regarding non-invasive remote patient monitoring in the context of COVID-19 reflects the stark reality of an unprecedented challenge we now face:

Healthcare as we currently practice it is not equipped to manage this massive problem.

Our healthcare infrastructure faces some enormous challenges related to this crisis:

  • Our frontline healthcare providers face high risk of infection given their daily exposure to COVID-19. This risk increases with the need for repeated patient contacts that are required to monitor patients in the traditional healthcare paradigm. When we think of the downstream effects of losing providers who are exposed to and may contract this disease and are then unable to care for patients, the problems of health system management become almost incomprehensibly compounded.
  • The supply chain of personal protective equipment (PPE) needed to protect patients and caregivers will face increasing strain and shortages. Caring for patients with infectious conditions requires an immense amount of disposable PPE and we are already seeing shortages of supplies that are impacting testing and have begun to impact care as the number of COVID-19 patients rise. This will exacerbate the risk to providers and their patients.
  • Our brick and mortar healthcare capacity will soon be overwhelmed with COVID-19 patients and we don’t have enough clinicians to take care of everybody. If Italy is any guide, and current trajectories indicate the U.S. should be prepared for this outcome, there are simply not enough hospital beds to care for the likely influx of COVID-19 patients as well as the rest of the population.

As the number of COVID-19 patients rapidly increases, the healthcare system must still care for the millions of chronically ill patients who require care and are at risk of clinical deterioration. The influx of COVID-19 patients overwhelm an already overworked healthcare system, forcing an almost unthinkable need to triage finite healthcare services and resources. Without bold and innovate thinking we will not be able to adequately address the challenges facing us in the management of our at-risk patient populations that includes the millions and millions of people with, for example, heart failure, pulmonary disease, and cancer. A quick review of the issues outlined above yield a logical common denominator:

In addition to the measures already being taken, there is an imperative to quickly and aggressively add continuous remote patient monitoring (CRPM) to our disease management paradigm.

Before the advent of clinical grade wearable sensors, the term “remote monitoring” was used to describe simple daily spot check vital signs like blood pressure or weight. This low data density offers limited utility in supporting the management of these at-risk patient populations.

However, over the past 5 years, there has been a quantum leap in our ability to continuously stream data from wearable sensors and apply artificial intelligence (AI) algorithms to provide clinicians with multi-dimensional clinical insight in near real time.

More specifically, we can now continuously stream every breath, every heartbeat, and every subtle movement using clinical grade disposable biosensors. This data set can then be combined with advanced AI-driven analytics that do the heavy lifting of data analysis to identify clinically relevant changes in health. From here, automated alerts to clinicians indicate who within the monitored population require medical attention. Such analytics help free the clinician from the need to manually process mountains of biosensor data and empowers them with the information they need to proactively care for deteriorating patients using subtle, often subclinical physiologic changes. Armed with this type of insight, clinicians can direct resources in a more informed, rational, effective and efficient way.

Continuous remote patient monitoring (cRPM) has enormous potential to help address the challenges outlined above in managing the current COVID-19 crisis and other resource challenges:

  • More clinical insight, less viral exposure. By applying AI to continuously streaming clinical-grade vitals from wearable sensors, clinicians achieve unprecedented clinical insight without direct patient contact. This has implications for both outpatient and inpatient care, where existing telemetry capacity is insufficient to address the surge demands being brought on by COVID-19.
  • Managing more patients: By leveraging artificial intelligence we have the opportunity to lean on machines to process more information than humans can…by a lot. This processing power is ideally suited to make sense of data streaming from wearable sensors and there is an unprecedented opportunity to use AI to provide a scalable way for clinicians to manage large at-risk patient populations.
  • Do more with less: By providing a continuous way to remotely monitor the status of patients, providers can minimize the frequency of taking vital signs, thereby limiting staff exposure, the risk of staff spreading disease to other patients, and the use of limited PPE.
  • Keep eyes on the at-risk non-COVID-19 patients. Prior to this crisis, our healthcare system was already operating close to capacity… these patients are not going away. The issue is that there is now nowhere for them to go. Continuous remote monitoring offers clinicians a tool they can use to ensure their non-COVID at risk populations don’t fall through the cracks.

In practical terms, how can this be applied? There are several use cases where hospitals and health systems should explore using AI-enhanced continuous remote monitoring to improve care, protect patients and staff, and conserve resource utilization:

  • Confirmed or suspected COVID-19 patients who are high risk (e.g., >50 years old with one or more co-morbidities) but still sub-acute. This would allow clinicians to effectively monitor patients who are self-quarantined at home.
  • Hospitalized COVID-19 patients who are in a general hospital bed where repeated staff interactions with patients puts caregivers at risk, supplies of disposable PPE become limited or traditional telemetry is not available.
  • High-risk populations (heart failure, COPD, chemotherapy) who are clinically fragile and need care, but can no longer access traditional or in-clinic resources due to lack of availability or risk of exposure during accessing the traditional healthcare system.

The current crisis requires that we think creatively in how we increase clinical capacity and care for at-risk patients. To be clear, there is no one magic solution or silver bullet. But the clinical world should know that, with AI and clinical grade wearables, they have an enormously powerful tool at their disposal. Big problems call for bold solutions.

Stephen L. Ondra, MD, is a seniors advisor to physIQ, Inc., and a board-certified neurological surgeon and national leader in medicine, medical policy, health information technology, and innovation. Dr. Ondra has held executive-level roles in health care reform, complex health care delivery systems, academe, and as member of the Obama administration. He previously served as the Senior Vice President and Enterprise Chief Medical Officer at the Health Care Service Corporation.